CLIN EXP RHEUM 28/4

Clinical and Experimental Rheumatology 2010; 28: 576-583.

Paediatric rheumatology

Condylar lesions in relation to mandibular growth in untreated and intra-articular corticosteroid-treated experimental

temporomandibular joint arthritisP. Stoustrup1, K.D. Kristensen1, A. Küseler1, J. Gelineck2, P.M. Cattaneo1,

T.K. Pedersen1, T. Herlin3

1Department of Orthodontics, 2Department of Radiology, and 3Paediatric Rheumatology Clinic, University of Aarhus, Denmark.

AbstractObjective

To evaluate condylar lesions in relation to mandibular growth in experimental temporomandibular joint (TMJ) arthritis and to assess the outcome of treating this condition with repeated intra-articular corticosteroid injections (IACIs).

MethodsForty-two 10-week-old rabbits were randomly divided into four groups. Seven animals served as controls. Experimental TMJ arthritis was induced in five animals which received intra-articular TMJ saline injections. Fifteen animals had TMJ arthritis induced and were left untreated and 15 animals had TMJ arthritis induced and were treated with IACIs one week

after each TMJ antigen-challenge procedure.Inter-group growth differences were evaluated from head computerised tomography scans taken at the time of arthritis induction and 12 weeks later. The variables assessed were: progression of condylar lesions (erosions/flattening/osteo-

phytes), mandibular bone volume changes, condylar and sagittal ramus growth.

ResultsNo inter-group differences in the progression of condylar lesions were observed despite reduced mandibular growth in all

three experimental groups. The most pronounced unfavourable mandibular growth alterations were observed in the corticosteroid-treated arthritis animals.

ConclusionNo evidence was found in support of a relation between reduced mandibular growth and condylar lesions. We propose that:

1) condylar lesions are not the only causative factor of reduced mandibular growth in experimental TMJ arthritis, and 2) repeated IACIs have a very unfavourable impact on mandibular growth in experimental TMJ arthritis – treatment is more

detrimental to mandibular growth than the TMJ arthritis itself.

Key wordsMandibular condyle, growth, intra-articular injections, steroids, experimental arthritis, temporomandibular joint,

bone resorption

577

PAEDIATRIC RHEUMATOLOGYCondylar lesions and mandibular growth / P. Stoustrup et al.

Peter Stoustrup, DDS, PhD Kasper D. Kristensen, DDS, PhDAnnelise Küseler, DDS, PhD John Gelineck, MD Paolo M. Cattaneo, MSc, PhD Thomas K. Pedersen, DDS, PhDTroels Herlin, MD, Professor.Please address correspondence and reprint requests to: Dr Peter Stoustrup, Department of Orthodontics, University of Aarhus, Vennelyst Boulevard 9, DK-8000 Aarhus C, Denmark. E-mail: [email protected] Grant supporters was received from the Danish Rheumatic Association, the Gangsted Foundation, Aarhus University Research Foundation and the Danish Dental Association.Received on September 21, 2009; accepted in revised form on March 11, 2010.© Copyright CLINICAL AND EXPERIMENTAL RHEUMATOLOGY 2010.

Competing interests: none declared.

IntroductionTemporomandibular joint (TMJ) arthri-tis is known to alter the mandibular de-velopment in children diagnosed with juvenile idiopathic arthritis (JIA) (1-5). It affects the condylar cartilage that forms the basis for the endochondral ossification process partly responsible for the mandibular growth (6). The lo-cation of this growth zone within the TMJ is an unique characteristic of this joint. Resultant abnormalities include condylar lesions, an unfavourable pos-terior mandibular rotation pattern, mi-crognathia, malocclusion with an an-terior open bite, altered joint and mus-cular function occasionally associated with pain (2, 7-9). Condylar lesions (erosions/flattening/osteophytes) have been regarded as the primary contribu-tory cause to the mandibular growth deviations in JIA patients with TMJ arthritis. Today, determination of the presence of TMJ inflammation as well as the long-term prognosis of man-dibular growth when TMJ arthritis has been diagnosed are therefore based on the presence of condylar lesions. The extent of these lesions is believed to play a significant role in the mandibu-lar and facial morphological deviations seen in JIA patients with TMJ arthritis (2). Naturally, this perception may be changed as new imaging methods, such as MRIs, gain a foothold in the diagno-sis of early TMJ arthritis in JIA. In JIA patients with TMJ arthritis, treat-ment mainly comprises orthopaedic procedures and corrective orthognatic surgery when condylar lesions are present and mandibular growth distur-bances have occurred (10, 11). Obvious advantages would therefore be gained if growth abnormalities and condylar lesions could be minimised or totally avoided through early aggressive in-tervention targeting the inflammatory changes in the joint. Intra-articular TMJ corticosteroid injections (IACI) have shown good results in terms of symp-tomatic relief and functional improve-ments in JIA patients (12-15). However, recommending IACI directly into the TMJs of growing JIA patients should await a deeper understanding of how the mandible and the lower face devel-op under the influence of TMJ arthritis.

Besides their anti-inflammatory and im-munosuppressive effect, corticosteroids are also known to suppress endochon-dral bone formation (16) which could have a most unfavourable long-term impact on mandibular growth because of the superficial position of the man-dibular growth plate. This aspect, yet to be resolved, has not been addressed in previous studies investigating IACIs in JIA patients with TMJ arthritis. The aim of the present experimental study was therefore to evaluate the follow-ing parameters in growing rabbits with untreated and IACI-treated antigen-in-duced TMJ arthritis: 1) The progression of condylar lesions as a reflection of the relationship between condylar le-sions and mandibular growth. 2) The mandibular bone volume changes dur-ing the growth period observed. 3) The mandibular growth in specific mandib-ular areas of interest as assessed from 3-dimensional superimpositions.

Material and methodsAnimals, arthritis induction and interventionsAll animal procedures were per-formed in accordance with a protocol approved by the Danish ethical com-mittee for animal welfare. The mate-rial comprised 10-week old female New Zealand white rabbits (n=42) (Oryctolagus cuniculus) housed at the animal facilities, Aarhus University, Denmark, with free access to food and water. Animal welfare was monitored by daily evaluation of food and wa-ter intake. Prior to arrival, all animals were randomly assigned to four groups by block randomisation (Fig. 1); con-trol group (n=7), placebo group (n=5), an untreated arthritis group (n=15) and a corticosteroid-treated arthritis group (n=15). All animals in the untreated ar-thritis group and in the corticosteroid-treated arthritis group had TMJ arthritis induced according to method described by Kapila et al. (17) and illustrated in Fig. 1. From age 10 weeks to age 14 weeks all animals in the three experi-mental groups were subcutaneously pre-sensitised towards the antigen ovalbumin (Sigma Chemicals) togeth-er with Freunds incomplete/complete adjuvant (IFA, Sigma Chemicals). At

578

PAEDIATRIC RHEUMATOLOGY Condylar lesions and mandibular growth / P. Stoustrup et al.

age 14, 17, 20 and 23 weeks, all ani-mals in the untreated arthritis group and the corticosteroid-treated arthritis group were injected with 0.1 ml 10 mg/ml ovalbumin in each TMJ to induce the experimental TMJ arthritis. In the same weeks the animals in the placebo group were injected with 0.1 ml saline in each TMJ. One week after each of the antigen-challenge procedures, the animals in the corticosteroid-treated arthritis group received 0.1 ml of 20 mg/ml Triamcinolone Hexacetonide (Lederspan®) in each TMJ (at age 15, 18, 21, 24 weeks). A total of ten rabbits were lost during the antigen challenge procedures due to anaphylactic shock. The lost animals were from the un-treated arthritis group (7 animals) and from the corticosteroid-treated arthritis group (3 animals). No animal was lost during the i.a. TMJ injections of corti-costeroid or saline (Fig. 1). At T2, the weights of all animals were obtained.Two sets of full-head computerised tomographic scans (CT scan) (Philips/Mx8000 IDT 16) were carried out at the time of arthritis induction when the ani-mals reached the age of 14 weeks (T1) and the age of 26 weeks (T2) which gave a 12-week observation period. Prior to the first CT scan session at T1, all ani-mals had two 1.0 x 0.33 mm tantalum implants inserted into the right side of the mandible. The CT-scans were used for the evaluation of; 1) progression of condylar lesions; 2) mandibular bone volume changes; 3) evaluation of con-dylar and sagittal growth on 3-dimen-sional superimpositions.

Condylar lesionsCondylar lesions were scored in a blinded fashion in accordance with a semi-quantitative modified Rohlin and Petersson scale on regular 2-dimension-al cuts together with 3-dimensional re-constructions of the mandibles (18). A condylar lesion score was assessed for each TMJ at T1 and T2. Each condyle was assessed with a lesion score from 0-5 as defined in Table I. A total con-dylar lesion progression score per TMJ was assessed (score at T2 minus score at T1 for each condyle), and for each animal a total score of the right and left joint was calculated to evaluate the

total TMJ lesion progression score per animal. The total TMJ lesion progres-sion score per animal was used in the statistical analysis of the test results to meet the requirements of independence in the statistical tests used. As an ex-ception to this, only the right side pro-gression score values were used in the evaluation of the correlation between the right side condylar growth and the condylar lesion progression scores, so that a direct correlation from the same side was obtained.

Mandibular segmentationsOn all CT-images the mandibles were manually separated from the base of the cranium using the software pro-gram Mimics (Mimics 10.1, Material-ise, Leuven, Belgium). The segmenta-tions were assessed in a blinded fashion before group assignment was revealed. Additionally, evaluation of the intra-observer variance in relation to the creation of the manual segmentation procedures was performed by ten ran-domly chosen duplicate segmentations.

Fig. 1. New Zealand White Rabbits (n=42) age 10 weeks, were randomly divided into four groups. †number of animals lost before the second computed tomographic scan session; thus data from 32 animals were available for the final growth analysis at T2.

Table I. Condylar lesion progression score scale used for the semi-quantitative evaluation.

Grade 0 Normal joint Bony joint surface of the condyle with a convex and well defined outline

Grade 1 Slightly abdormal joint Single, minor changes interpreted as uncertain

Grade 2 Definitive early abnormality sign Definitive minor surface-changes such as erosions

Grade 3 Moderate destructive abnormality Erosions and local condylar changes such as a V-shaped tubercle

Grade 4 Severe destructive abnormality Extensive mediolateraly erosions of the condyle extending through a minimum of 1/3 of the total sagittal condylar dimension

Grade 5 Mutilating abnormality Total resorption of the condylar head with disappearance of the articular surface

579


Mandibular superimpositionsEach 3-dimensional mandibular T1 im-age was superimposed on the T2 image of the same animal. Superimpositions were created on the inserted implants and stable anatomical landmarks. Two anatomical variables were measured on the superimpositions for a quantita-tive evaluation of inter-group changes in mandibular development; condylar

growth (CG) and sagittal ramus growth (SRG) (Figures 2a and 2b). The animals were sacrificed at the age 26 weeks after the second CT scan session (T2). Implant operations, TMJ injections and euthanisation were carried out under general anaesthesia. All intra-articular TMJ injections and surgical procedures were carried out by trained specialists in a blinded fashion. The animals that did

not complete the full length of the study were excluded from the final analysis. Thus, data from 32 animals were availa-ble for the final analysis. Only statistical analyses pre-specified in the trial proto-col were performed, except for the two variables CG and SRG measured on the superimpositions.

Statistics, condylar lesionsThe reproducibility of the condylar le-sion progression score per TMJ, ex-pressed in Kappa values, was evaluated by duplicate assessments two months apart in a blinded fashion by the same observer (PS). A Wilcoxon matched-paired rank-test was used for the evalu-ation of intra-group differences between T1 and T2. A non-parametric Kruskall-Wallis test was used for the evaluation of inter-group differences in the TMJ le-sion progression score per animal at T2.

Statistics, mandibular segmentationsIntra-observer variance in the man-dibular segmentation process was evaluated according to the guidelines for duplicate measurements described by Bland and Altmann (19). Absolute mandibular inter-group growth dif-ferences at baseline (T1) were evalu-ated by an ANOVA one-way analysis. Evaluation of absolute mandibular in-tra-group growth between T1 and T2 was carried out by paired Student’s t-test. Inter-group differences in relation to relative mandibular bone volume changes between the T1 and T2 were evaluated by a one-way ANOVA anal-ysis with two-paired t-tests serving as post-ANOVA tests.

Statistics, mandibular superimpositionsInter-group differences in condylar growth and sagittal ramus growth were evaluated by one-way-ANOVA tests with two-paired t-tests serving as post-ANOVA tests. The need for Bonferroni corrections to avoid mass significance in the post-ANOVA inter-group t-tests was evaluated. However, to avoid type-1 er-ror due to the correlation between the variables evaluated in each animal, this correction was considered too strong and therefore was not implemented in the statistical analysis. A p-value of <0.05 was considered significant.

Fig. 2. Superimposition of a control group animal. The red parts represent the mandible at age 14 weeks (T1) and the yellow parts represent the mandible at age 26 weeks (T2). Mandibular scans from each animal at T1 and T2 were superimposed on implants and stable anatomical landmarks. The con-dylar growth distance is illustrated together with the sagittal ramus growth distance. Both of these variables were measured on all superimpositions.

Fig. 3. a) Total lesion progression score per animal between T1 and T2. The lesion score of the right and the left TMJ were added for a total lesion progression score per animal. Probability of in-ter-group differences p=0.37. Notice that 3 animals obtained a negative condylar lesion progression score indicating improvements in the condylar dysmorphology between T1 and T2. b) Mean relative bone-volume changes (%) in each group during the 12 weeks of growth observed. The mean value is depicted with a straight line. c) Inter-group differences in absolute condylar growth; mean value is depicted with a straight line. d) Inter-group differences in absolute sagittal ramus growth; mean value is depicted with a straight line.* indicates a significance level of p<0.05 and ** indicates a significance level of p<0.001.

580


ResultsNo inter-group differences in the ani-mals’ weights were observed at T2 (ANOVA, probability of inter-group differences p=0.29).

Condylar lesionsKappa statistics from duplicate as-sessments at T2 showed an accept-able reproducibility of 78% per TMJ. No inter-group differences at T1 were observed. The condylar lesion progres-sion scores for all animals in each of the groups are illustrated in Fig. 3a. No intra-group differences between T1 and T2 were observed. Additionally, no in-ter-group differences in the progression of the condylar lesions were observed.

Mandibular segmentationsAll variables were evaluated for normal distribution. Duplicate manual man-dibular segmentations showed an intra-individual variation coefficient of 0.99 in relation to the bone volume variable. This variation was found acceptable af-ter graphical evaluation (19). At T1, we observed no significant inter-group dif-ferences in bone volume changes. Sig-nificant mandibular intra-group bone volume changes were seen between T1 and T2 in all four groups. Table II depicts T1 and T2 bone volume proper-ties (mean ± SD) for each of the groups. At T2, significantly reduced relative mandibular bone volume enlargements were observed in the placebo group (16.7%, 95-CI: 14.9–18.4), the arthritis group (14.2%, 95-CI: 11.9–16.6) and the corticosteroid group (11.1%, 95-CI: 8.8–13.3) compared with the control group (21.8%, 95-CI: 19.2–25) (Fig. 3b). During the 12 weeks of observa-tion, the significantly smallest relative bone volume changes were observed in the corticosteroid-treated animals.

Mandibular superimpositionsFigures 3c and 3d depict inter-group differences in absolute condylar growth and sagittal ramus growth. The condylar growth was significantly reduced in the corticosteroid-treated animals (1.4 mm 95-CI: 0.9–1.9) compared with the con-trol group (3.7 mm, 95-CI: 3.2–4.3), the placebo group (3.2 mm, 95-CI: 2.3–4.1) and the arthritis group (2.7 mm, 95-CI:

1.8–3.5). When examining the sagittal ramus growth (SRG), this variable was significantly reduced in the arthritis ani-mals (4.0 mm, 95-CI: 3.5–4.6) and in the corticosteroid-treated animals (3.8 mm, 95-CI: 3.4–4.1) compared with the control group animals (4.9 mm, 95-CI: 4.3–5.4).

Progression of condylar lesions in relation to mandibular growthA direct comparison between the con-dylar progression scores and the growth variables relative volume changes and condylar growth are illustrated in Fig. 4.

Discussion The primary objective of this study was to evaluate the relation between condy-lar lesions and mandibular growth in untreated and IACI-treated experimen-tal TMJ arthritis. The findings of this study were the following. Firstly, the mandibular growth was reduced in all three experimental groups compared with controls in terms of relative bone

volume changes (Fig. 3b). Secondly, reduced mandibular growth was ob-served in the untreated arthritis group compared with the control group which supports the perception that arthri-tis activity per se reduces mandibular growth. Thirdly, the most pronounced unfavorable mandibular growth altera-tions were observed in the corticoster-oid-treated arthritis animals. Fourthly, inter-group differences in mandibular growth occurred without any inter-group differences in the progression of condylar lesion scores. No relationship between the severity of the condylar le-sion progression scores and the reduc-tions in the mandibular growth vari-ables was found. Fifthly, experimental TMJ arthritis has an extra-articular consequence such as reduced sagittal ramus growth which is seen in the un-treated arthritis group and the corticos-teroid-treated arthritis group. Extant research has been debating the effect of condylar lesions on mandibu-lar growth for decades (2, 20, 21). In

Table II. Age 14 weeks (T1) and age 26 weeks (T2) data with reference to mandibular bone-volume, condylar growth and sagittal ramus growth between T1 and T2.

Group Mean Mean Mean bone-volume changes condylar growth sagittal (T2-T1) (mm) + (SD) ramus growth (mm3) ± (SD) (mm) ± (SD)

Control group 1957 ± (202) 3.71 ± (0.5) 4.85 ± (0.21)

Placebo group 1479 ± (128) 3.23 ± (0.32) 4.32 ± (0.13)

Arthritis group 1241 ± (246) 2.66 ± (0.22) 4.01 ± (0.24)

Corticosteroid group 1008 ± (359) 1.41 ± (0.239 3.77 ± (0.16)

Fig. 4. a) Lesion progression score (per animal) in relation to mandibular relative bone volume chang-es. b) Right side lesion progression score in relation to the right side condylar growth. Notice, that only the right side lesion progression scores were used in this evaluation, so that a direct correlation between condylar growth and the lesion progression score from the same side was evaluated. A visual inspec-tion of the two graphs as well as their regression lines (correlation r2-values) indicate no relationship between the severity of the condylar lesion progression scores and the reductions in mandibular growth observed in this experimental temporomandibular joint arthritis model.

581


the clinical evaluation of the presence and severity of TMJ arthritis in JIA children, many pediatric rheumatolo-gists believe that condylar lesions are the primary factor causing mandibu-lar alterations and growth reduction (7, 22). The present findings allow us to question this claim. A contrast-en-hanced MRI study by Küseler et al. (1998) documented the existence of inflammatory changes in the TMJ in JIA patients before they could be re-vealed by conventional radiographics and clinical symptoms (8, 23). Recent cohort studies from Twilt et al. have demonstrated mandibular retrognathia and posterior rotations in JIA patients without detectable condylar alterations on orthopantomograms. Twilt et al. also observed improvement and even regeneration of these TMJ alterations over time (5, 24). Argyropoulou et al. found that an abnormal condyle and articular eminence was correlated with a long inflammatory activity duration (25). In accordance with these findings, our results allow us to hypothesise that condylar alterations are the product of manifest and long-term inflammatory TMJ changes rather than an initial oc-curring condition in early TMJ arthritis. Our results also invite us to suggest that TMJ inflammation may critically affect the endochondral condylar ossification even in the absence of condylar lesions. Future clarification of this is essential and of utmost importance for the diag-nostic and therapeutic understanding of mandibular growth alterations in JIA children. In previous experimental studies, we have shown that IACIs reduce the in-flammatory response in growing rab-bits challenged with antigen-induced TMJ arthritis (26, 27). Arabshahi et al. (2005) obtained good clinical results by treating TMJ arthritis in JIA with a single IACI which induced significant pain resolution, improved joint func-tion and induced resolution of joint ef-fusion detected by MRI (12). Ringold et al. (2008) have published data dem-onstrating improved TMJ function fol-lowing the initial IACI, but a minimal response to subsequent injections (15). These studies support symptomatic ef-fect of IACIs against TMJ arthritis in

JIA. However, these studies have gen-erally seen neglected the important issue of a potential negative effect of the injected corticosteroids on the con-dylar cartilage and mandibular growth described in experimental studies (28-30): the potential long-term side-effects on mandibular growth may outweigh the beneficial short-term functional improvements of IACIs against TMJ inflammation. Unlike other joints, the condylar cartilage is an intra-articular active site for endochondral bone for-mation of utmost importance to man-dibular development (6). The mandibu-lar growth plate is situated just beneath the fibrocartilage of the condylar head, and this study supports the perception that this position makes it vulnerable to inflammation within the TMJ and that the disease activity itself interferes with mandibular growth. Our results suggest that this localisation makes the condylar cartilage even more vul-nerable to TMJ arthritis treated with repeated IACIs – even more than if the TMJ arthritis was left untreated. Of particular importance is that man-dibular growth was most reduced in the corticosteroid-treated arthritis animals, which indicates that corticosteroid-in-duced mandibular growth is hampered despite the beneficial effect of IACIs on TMJ inflammation described by Kristensen et al. (27). A number of limitations in this study need further considerations. The cor-ticosteroid-treated arthritis animals re-ceived more intra-articular injections per animal than any of the other group. However, we have previously shown that intra-articular TMJ saline injections per se do not result in any inflammatory or destructive structural changes in the joint (27). We therefore do not attach particular importance to the fact that the corticosteroid-treated animals re-ceived a higher number of actual TMJ injections per joint than the animals that received no corticosteroids.Significantly reduced relative man-dibular bone volume changes were observed at T2 in the placebo group compared with the control group. We ascribe this finding to the fact that the placebo animals were systemically pre-sensitised from age 10 to age 14

weeks, which is at the end of the rab-bit growth spurt that terminates around age 16 weeks (31). In contrary to the three experimental groups, the control animals were not pre-sensitised. We hypothesise that the bone volume dif-ference between the placebo and the control animals at T2 are due to the rather harsh systemic ovalbumin pre-sensitisation given to the experimental groups. This procedure may have cre-ated a general inflammatory condition influencing the well-being of the place-bo animals causing reduced mandibu-lar bone volume changes in this group. Ideally, the control group should have been subjected to the same systemic ovalbumin presensitisation procedures as the three experimental groups. How-ever, it is important to emphasise that in relation to the important variable, condylar growth, no differences be-tween the control animals and the pla-cebo treated animals were observed at T2. Based on the context of the present study the condylar growth variable is a more descriptive variable than the man-dibular bone volume because it directly describes the product of the single re-gional growth centre situated within the TMJs on the condylar head. In contrast, the final mandibular bone volume is the product of several regional mandibular growth areas, all potentially influenced by extra-articular occurrences such as systemic pre-sensitisation, resulting in reduced mandibular bone volume. A substantial limitation to this study is the great loss of animals (n=7) from the untreated arthritis group before the final analysis. Together with the lost corticosteroid treated animals (n=3), these were lost due to an anaphylactic reaction a few minutes after the ovalbu-min was injected into the TMJs. If these animals were lost because they were most severely affected by the experi-mental arthritis, it could be anticipated that they would also have the most pro-nounced mandibular growth deviations. Yet, the study design does not allow us to conclude on this. When a study consists of a reduced amount of animals it is always a limita-tion when further loss of animals occur and great intra-group result deviations are seen. In this study, great deviations

582


were seen within the condylar growth variable. However, these deviation very much reflect the great histological in-tra-group deviations seen in this study presented by Kristensen et al. (27). In this study the arthritis was intro-duced at age 14 weeks in conformity with the Danish regulations on animal welfare. For the evaluation of an early treatment intervention against inflam-matory changes in the TMJs in young subjects, it would have been preferable to initiate the trial in even younger ani-mals in which case we would expect even greater inter-group differences in mandibular disturbances. Evaluation of mandibular catch-up growth after ex-piry of the effect of the corticosteroid on the condylar endochondral ossifica-tion is not possible due to the design of the study. In rabbits, approximately 90% of the overall mature sagittal and transversal mandibular growth is achieved within the sixteenth week and has ceased at age 26 weeks (31) so it is questionable if growth could be ex-pected after the 26th week. Applying findings obtained in experi-mental studies to man requires care-ful consideration. Even so, in terms of altered mandibular growth caused by TMJ arthritis, this experimental study does allow us to propose two interesting statements essential for future clinical investigations: 1) TMJ inflammation interferes with condylar growth and mandibular development and growth may be reduced even in the absence of condylar alterations; 2) Although treat-ment with IACI for TMJ arthritis has shown clinical improvement (12, 15) none of the human studies, so far, have described the long-term effect of IACI on mandibular growth in children. It is still unknown whether the use of re-peated IACIs for TMJ arthritis in chil-dren may reduce mandibular growth even more than if the arthritis is left untreated, as we have observed in this experimental study.To avoid potential mandibular growth reductions caused by TMJ inflamma-tion other pharmalogical approaches could be considered. Recently, we have shown that intervention with systemic etanercept monotherapy equivalent to the recommended human dose (0.8 mg/

kg weekly) has an anti-inflammatory ef-fect on the synovial tissues in the TMJ and furthers mandibular growth towards an original morphology in experimental TMJ arthritis in rabbits (32, 33). We believe that our results are of im-portance for clinicians treating children with TMJ arthritis as repeated IACIs against TMJ arthritis may have an un-expected negative effect on mandibu-lar growth. We therefore recommend a careful follow-up of patients with TMJ injections in clinical studies to evaluate the potential negative influence of IACIs on the mandibular growth in humans.

Reference 1. BILLIAU AD, HU Y, VERDONCK A, CARELS

C, WOUTERS C: Temporomandibular joint arthritis in juvenile idiopathic arthritis: prev-alence, clinical and radiological signs, and relation to dentofacial morphology. J Rheu-matol 2007; 34: 1925-33.

2. KJELLBERG H: Juvenile chronic arthritis. Dentofacial morphology, growth, mandibu-lar function and orthodontic treatment. Swed Dent J Suppl 1995; 109: 1-56.

3. MERICLE PM, WILSON VK, MOORE TL et al.: Effects of polyarticular and pauciarticular onset juvenile rheumatoid arthritis on facial and mandibular growth. J Rheumatol 1996; 23: 159-65.

4. SIDIROPOULOU-CHATZIGIANNI S, PAPADO-POULOS MA, KOLOKITHAS G: Dentoskeletal morphology in children with juvenile idi-opathic arthritis compared with healthy chil-dren. J Orthod 2001; 28: 53-8.

5. TWILT M, SCHULTEN AJ, NICOLAAS P, DULG-ER A, SUIJLEKOM-SMIT LW: Facioskeletal changes in children with juvenile idiopathic arthritis. Ann Rheum Dis 2006; 65: 823-5.

6. ENLOW DH, HANS MG, EDITORS: Growth of the mandible. Essential of facial growth. Philadelphia, Saunders, 1996: 57-78.

7. PEDERSEN TK, JENSEN JJ, MELSEN B, HER-LIN T: Resorption of the temporomandibular condylar bone according to subtypes of juve-nile chronic arthritis. J Rheumatol 2001; 28: 2109-15.

8. PEDERSEN TK, KUSELER A, GELINECK J, HERLIN T: A prospective study of magnetic resonance and radiographic imaging in rela-tion to symptoms and clinical findings of the temporomandibular joint in children with ju-venile idiopathic arthritis. J Rheumatol 2008; 35: 1668-75.

9. STABRUN AE: Impaired mandibular growth and micrognathic development in children with juvenile rheumatoid arthritis. A longi-tudinal study of lateral cephalographs. Eur J Orthod 1991; 13: 423-34.

10. PEDERSEN TK, GRONHOJ J, MELSEN B, HER-LIN T: Condylar condition and mandibular growth during early functional treatment of children with juvenile chronic arthritis. Eur J Ortho 1995; 17: 385-94.

11. PEDERSEN TK: Clinical aspects of ortho-dontic treatment for children with juvenile

chronic arthritis. Acta Odontol Scand 1998; 56: 366-8.

13. ARABSHAHI B, DEWITT EM, CAHILL AM et al.: Utility of corticosteroid injections for temporomandibular arthritis in children with juvenile idiopathic arthritis. Arthritis Rheum 2005; 52: 3563-9.

14. BEUKELMAN T, GUEVARA JP, ALBERT DA, SHERRY DD, BURNHAM JM: Usage of in-tra-articular corticosteroid injections for the treatment of juvenile idiopathic arthritis: a survey of pediatric rheumatologists in the United States and Canada. Clin Exp Rheu-matol 2008; 26: 700-3.

15. WEISS PF, ARABSHAHI B, JOHNSON A et al.:High prevalence of temporomandibular joint arthritis at disease onset in children with juvenile idiopathic arthritis, as detected by magnetic resonance imaging but not by ultrasound. Arthritis Rheum 2008; 58: 1189-96.

15. RINGOLD S, TORGERSON TR, EGBERT MA, WALLACE CA: Intraarticular corticosteroid injections of the temporomandibular joint in juvenile idiopathic arthritis. J Rheumatol 2008; 35: 1157-64.

16. MACRAE VE, FARQUHARSON C, AHMED SF: The pathophysiology of the growth plate in juvenile idiopathic arthritis. Rheumatology (Oxford) 2006; 45: 11-9.

17. KAPILA S, LEE C, TAVAKKOLI JOU MR, MILLER AJ, RICHARDS DW: Development and histologic characterizations of an ani-mal model of antigen-induced arthritis of the juvenile rabbit temporomandibular joint. J Dent Res 1995; 74: 1870-9.

18. HU YS, SCHNEIDERMAN ED: The temporo-mandibular joint in juvenile rheumatoid ar-thritis: 1. Computed tomographic findings. Pediatr Dent 1995; 17: 46-53.

19. BLAND JM, ALTMAN DG: Statistical methods for assessing agreement between two meth-ods of clinical measurement. Lancet 1986; 1: 307-10.

20. LARHEIM TA, HAANAES HR, RUUD AF: Man-dibular growth, temporomandibular joint changes and dental occlusion in juvenile rheumatoid arthritis. A 17-year follow-up study. Scand J Rheumatol 1981; 10: 225-33.

21. STABRUN AE, LARHEIM TA, HOYERAAL HM, ROSLER M: Reduced mandibular dimensions and asymmetry in juvenile rheumatoid ar-thritis. Pathogenetic factors. Arthritis Rheum 1988; 31: 602-11.

22. TWILT M, VAN DER GE, MOBERS SM, TEN CATE R, SUIJLEKOM-SMIT LW: Abrupt con-dylar destruction of the mandibula in juvenile idiopathic arthritis. Ann Rheum Dis 2003; 62: 366-7.

23. KUSELER A, PEDERSEN TK, HERLIN T, GE-LINECK J: Contrast enhanced magnetic reso-nance imaging as a method to diagnose early inflammatory changes in the temporoman-dibular joint in children with juvenile chronic arthritis. J Rheumatol 1998; 25: 1406-12.

24. TWILT M, SCHULTEN AJ, VERSCHURE F, WISSE L, PRAHL-ANDERSEN B, SUIJLEKOM-SMIT LW: Long-term followup of temporo-mandibular joint involvement in juvenile idiopathic arthritis. Arthritis Rheum 2008; 59: 546-52.

25. ARGYROPOULOU MI, MARGARITI PN,

583


KARALI A et al.: Temporomandibular joint involvement in juvenile idiopathic arthritis: clinical predictors of magnetic resonance im-aging signs. Eur Radiol 2009; 19: 693-700.

26. KUSELER A, PEDERSEN TK, BARLACH J et al.: Contrast-enhanced MRI compared to histological findings in the temporomandibu-lar joint of antigen-induced arthritis in young rabbits. Clin Exp Rheumatol 2004; 22: 441-6.

27. KRISTENSEN KD, STOUSTRUP P, KUSELER A et al.: Quantitative histological changes of repeated antigen-induced arthritis in the tem-poromandibular joints of rabbits treated with intra-articular corticosteroid. J Oral Pathol Med 2008; 37: 437-44.

28. JUX C, LEIBER K, HUGEL U et al.: Dexa-

methasone impairs growth hormone (GH)-stimulated growth by suppression of local insulin-like growth factor (IGF)-I production and expression of GH- and IGF-I-receptor in cultured rat chondrocytes. Endocrinology 1998; 139: 3296-305.

29. SILBERMANN M, MAOR G: Mandibular growth retardation in corticosteroid-treated juvenile mice. Anat Rec 1979; 194: 355-67.

30. SILVESTRINI G, BALLANTI P, PATACCHI-OLI FR et al.: Evaluation of apoptosis and the glucocorticoid receptor in the cartilage growth plate and metaphyseal bone cells of rats after high-dose treatment with corticos-terone. Bone 2000; 26: 33-42..

31. MASOUD I, SHAPIRO F, MOSES A: Longi-tudinal roentgencephalometric study of the

growth of the New Zealand white rabbit: cu-mulative and biweekly incremental growth rates for skull and mandible. J Craniofac Genet Dev Biol 1986; 6: 259-87.

32. KRISTENSEN KD, STOUSTRUP P, KUSELER A et al.: Intra-articular vs. systemic administra-tion of etanercept in antigen-induced arthritis in the temporomandibular point. Part I: his-tological effects. Pediatr Rheumatol Online J 2009; 7: 5.

33. STOUSTRUP P, KRISTENSEN KD, KUSELER A, PEDERSEN TK, GELINECK J, HERLIN T: Intra-articular vs. systemic administration of etanercept in antigen-induced arthritis in the temporomandibular joint. Part II: mandibular growth. Pediatr Rheumatol Online J 2009; 7: 6.

Documents

CLIN EXP RHEUM 28/4